During the initial funding period, members of the Program Project have explored the biological basis of fluorescence and reflectance spectroscopy of the cervix, developed a device (the Fast EEM) for collecting spectral information from point sources in the cervix, and begun clinical trials evaluating the ability of the device to detect cervical dysplasia. The Biomathematics and Informatics Core (BIC) has been vital to all these activities. The BIC created a database, implemented data quality assurance procedures, implemented data visualization software, investigated diagnostic classification algorithms, devised methods for variable selection, and analyzed data from a study of fluorescence spectroscopy over the menstrual cycle. In the new proposal, we will evaluate the Fast EEM device in a randomized prospective clinical trial. We will also begin evaluation of two new devices that extend the work begun with the Fast EEM. The Fast EEM records fluorescence intensities at a number of excitation emission wavelength pairs; the data are stored in an excitation emission matrix or EEM. The Fast EEM also measures reflectance intensities at a number of wavelengths and source-detector separations. The Multispectral Digital Colposcope (MDC) will collect reflectance and fluorescence data similar to that collected by the Fast EEM, but from the entire surface of the cervix. The Multispectral Optical Wand (MOW) will collect point source information similar to that collected by the Fast EEM, but at a greatly reduced number of wavelengths. When both devices are used in conjunction, with the MDC providing guidance for point source collection by the MOW, we hope these devices will achieve the sensitivity and specificity of colposcopically directed biopsy at reduced costs and greater satisfaction to patient, payer, and provider.